Skip to main content

John Deere acquires crop intelligence startup Orchard AI

John Deere has acquired San Francisco crop intelligence firm Orchard AI. Terms undisclosed. Its tractor AI cameras claim 95 per cent yield accuracy.

John Deere acquires crop intelligence startup Orchard AI

John Deere has acquired Orchard AI, a San Francisco based crop intelligence company that fits cameras and on board artificial intelligence to tractors and other farm vehicles. AgFunderNews reported the deal on 20 November and said the terms were not disclosed. The trade title AgNavigator reported it the same day, noting that the announcement lands days before Deere publishes its full year 2026 financial results. Deere's own announcement, headlined John Deere Acquires Orchard AI, uses that name for the business, while AgFunderNews refers to the company as Orchard Robotics.

The company was founded in 2022 by Charlie Wu, then a student at Cornell University. Its camera mounts on a vehicle already driving the block and captures millions of images as it goes, with the processing done on an on board computer rather than in the cloud. From those images it derives fruit counts, fruit size and colour, growth rates, canopy health, water stress, disease pressure and a yield estimate, all of which land in the company's FruitScope software. Wu told AgFunderNews that there are hundreds of systems in the field scanning tens of thousands of acres, with the ambition to reach hundreds of thousands.

The company started in apples and has since worked in wine grapes, table grapes, cherries, blueberries, almonds, pistachios and citrus, and is now moving into pomegranates, coffee and strawberries. Most of its customers are in the United States. Wu framed the business in unusually blunt terms: "We're not actually selling hardware. We're not even selling software. We're not even selling data." What growers buy, he said, is the outcome. On where the money shows up, he said "The biggest ROI is through reduction on labor and inputs," and that yields can be predicted with better than 95 per cent accuracy, which in turn drives labour, bin and storage planning.

For Deere the logic is integration rather than a new product line. Tim Marquis, Deere director of farm systems innovation, told AgFunderNews that "This connects crop-level intelligence with John Deere technology and equipment that can precisely execute the work in the field." AgNavigator reported that Deere can pull the software into its equipment offerings including the G5 in cabin display, and Justin Rose, president of the worldwide agriculture and turf division, small ag and turf, said in a statement that Deere wants to preserve the company's speed of innovation and its direct customer relationships while adding Deere's scale. The purchase continues a long run of buying: Tenna earlier in 2026, Sentera and GUSS Automation in 2025, Bear Flag Robotics in 2021 and Blue River Technology in 2017. Jorge Heraud, who sold Blue River to Deere in 2017, called the deal "further validation that large companies in agriculture are finding that actionable data is very valuable to farmers."

For founders building in specialty crops, the shape of this exit matters more than the price. Orchard AI sold four years after it was founded, with hundreds of units in commercial use rather than a pilot pipeline, and it sold to the company whose displays and sprayers its data ultimately has to reach. That is the pattern Deere has repeated since 2017: it buys the sensing and the decision layer, and the execution hardware stays its own. If your roadmap depends on reaching a grower through a Deere terminal, assume you are building towards either a deep integration agreement or an acquisition, and budget the engineering for both. The less comfortable corollary is that a sensing business with no credible route onto a major's machines has a much shorter buyer list, and the buyers who remain tend to pay in strategic patience rather than cash.

Source: AgFunderNews

John Deere acquires crop intelligence startup Orchard AI

The San Francisco company mounts AI cameras on tractors in orchards and vineyards and claims yield forecasts accurate to better than 95 per cent. Financial terms were not disclosed.

By Dirk Vandenhirtz

· AgTech AI · North America · 3 min

John Deere has acquired Orchard AI, a San Francisco based crop intelligence company that fits cameras and on board artificial intelligence to tractors and other farm vehicles. AgFunderNews reported the deal on 20 November and said the terms were not disclosed. The trade title AgNavigator reported it the same day, noting that the announcement lands days before Deere publishes its full year 2026 financial results. Deere's own announcement, headlined John Deere Acquires Orchard AI, uses that name for the business, while AgFunderNews refers to the company as Orchard Robotics.

The company was founded in 2022 by Charlie Wu, then a student at Cornell University. Its camera mounts on a vehicle already driving the block and captures millions of images as it goes, with the processing done on an on board computer rather than in the cloud. From those images it derives fruit counts, fruit size and colour, growth rates, canopy health, water stress, disease pressure and a yield estimate, all of which land in the company's FruitScope software. Wu told AgFunderNews that there are hundreds of systems in the field scanning tens of thousands of acres, with the ambition to reach hundreds of thousands.

The company started in apples and has since worked in wine grapes, table grapes, cherries, blueberries, almonds, pistachios and citrus, and is now moving into pomegranates, coffee and strawberries. Most of its customers are in the United States. Wu framed the business in unusually blunt terms: "We're not actually selling hardware. We're not even selling software. We're not even selling data." What growers buy, he said, is the outcome. On where the money shows up, he said "The biggest ROI is through reduction on labor and inputs," and that yields can be predicted with better than 95 per cent accuracy, which in turn drives labour, bin and storage planning.

For Deere the logic is integration rather than a new product line. Tim Marquis, Deere director of farm systems innovation, told AgFunderNews that "This connects crop-level intelligence with John Deere technology and equipment that can precisely execute the work in the field." AgNavigator reported that Deere can pull the software into its equipment offerings including the G5 in cabin display, and Justin Rose, president of the worldwide agriculture and turf division, small ag and turf, said in a statement that Deere wants to preserve the company's speed of innovation and its direct customer relationships while adding Deere's scale. The purchase continues a long run of buying: Tenna earlier in 2026, Sentera and GUSS Automation in 2025, Bear Flag Robotics in 2021 and Blue River Technology in 2017. Jorge Heraud, who sold Blue River to Deere in 2017, called the deal "further validation that large companies in agriculture are finding that actionable data is very valuable to farmers."

For founders building in specialty crops, the shape of this exit matters more than the price. Orchard AI sold four years after it was founded, with hundreds of units in commercial use rather than a pilot pipeline, and it sold to the company whose displays and sprayers its data ultimately has to reach. That is the pattern Deere has repeated since 2017: it buys the sensing and the decision layer, and the execution hardware stays its own. If your roadmap depends on reaching a grower through a Deere terminal, assume you are building towards either a deep integration agreement or an acquisition, and budget the engineering for both. The less comfortable corollary is that a sensing business with no credible route onto a major's machines has a much shorter buyer list, and the buyers who remain tend to pay in strategic patience rather than cash.

Source: AgFunderNews

Back to AgTech News